B-Cell Depleting Therapies: Fire-and-Forget or Treat-to-Target?

‘Standard’ dosing of rituximab

Rituximab dosing for renal indications was originally derived from the Rheumatology and Hemato-oncology literature. Conventionally, rheumatoid arthritis was treated with two rituximab doses, 1 g each, 2 weeks apart,¹ and lymphoma was treated with four rituximab doses, 375 mg/m² each, administered at weekly intervals.² Indeed, these are the doses currently recommended by the KDIGO Glomerular Diseases guideline (2021).³ However, it has been suggested that lower doses may be sufficient to achieve similar therapeutic efficacy.

Targeting B-cell depletion after rituximab

Ruggenenti et al. have long advocated for a single-dose regimen, with subsequent B-cell-guided re-dosing in membranous nephropathy (MN).⁴ They examined this strategy in 12 incident patients with idiopathic MN who received 375 mg/m² as a single dose, followed by repeat dosing only if B-cells≥5 cells/mm³ were detected in circulation. Of the 12 patients studied, only one required a second dose during the 1-year study. Despite this low cumulative dose, response rates were similar to historical controls who received the “full dose” of 375 mg/m² weekly for 4 weeks.⁵

In a separate investigation, healthy volunteers were administered rituximab infusions at 0.1, 0.3, or 1.0 mg/m² doses. This resulted in transient depletion of CD20 cells by a mean of 68%, 74%, and 97%, respectively. This demonstrated that, at least in healthy volunteers, ≤1% of the authorized rituximab dose was sufficient to deplete (albeit transiently) almost all circulating B lymphocytes.⁶ These data suggests that a lower rituximab dose is sufficient to deplete B-cells.

Defining depletion: Lessons from LUNAR

Arguably, the most informative insights into B-cell depletion as a therapeutic target come from the LUNAR trial. In this randomized double-blind phase III trial, patients with proliferative lupus nephritis treated with mycophenolate mofetil and corticosteroids were randomly assigned to receive add-on rituximab (1000 mg) or placebo on days 1, 15, 168, and 182.⁷ Routine flow cytometry analyses were performed to monitor B-cell depletion every 2 weeks for the first month, every 4 weeks until week 52, and then every 12 weeks until week 78. Although the trial found that add-on rituximab did not improve clinical outcomes after 1 year of treatment, significant insights can be gained from the subsequent post-hoc analysis of B-cell depletion.⁸

There is no clear consensus on the threshold that defines peripheral B-cell depletion. For example, “depletion” has variously been defined as a peripheral CD19 count of ≤1,⁹ ≤5,¹⁰ and even ≤20⁸ cells/mm³. The post-hoc analysis of LUNAR, however, evaluated a more stringent, exploratory definition of “complete peripheral depletion,” defined as a peripheral blood CD19 count of 0 cells/mm³. While all patients received the same dose of rituximab in LUNAR, interindividual variability was significant: 100% of participants depleted to ≤5 cells/mm³, but only 78% achieved complete peripheral depletion to 0 cells/mm³. Furthermore, while rituximab was administered on days 1, 15, 168, and 182, patients continued to progress to complete peripheral B-cell depletion for up to a year after the first rituximab dose. The median duration of complete peripheral depletion was 71 days.⁸

Most importantly, this analysis found that achieving complete peripheral depletion (i.e., CD19 count of 0 cells/mm³) conferred a greater likelihood of complete response at week 78. On the other hand, achieving CD19 counts of ≤20 cells/mm³ or ≤5 cells/mm³ (as per prior definitions of ‘depletion’) did not distinguish participants who achieved complete response from those who did not.⁸

The duration of depletion matters: Data from NOBILITY

Post-hoc analysis of LUNAR also demonstrated that the duration of sustained B-cell depletion was associated with a complete response at week 78.⁸ These findings appear compatible with those from NOBILITY, a phase 2 randomized controlled trial of patients with proliferative lupus nephritis that studied the humanized type II anti-CD20 monoclonal antibody, obinutuzumab. Patients receiving standard-of-care therapy with mycophenolate mofetil and corticosteroids were randomized to receive obinutuzumab (1000 mg) vs. placebo, on day 1, and weeks 2, 24, and 26. Improved renal responses were noted in the obinutuzumab arm at weeks 52 and 104.¹¹

The similarity in trial designs invites a comparison of the B-cell depleting ability of rituximab (in LUNAR) and obinutuzumab (in NOBILITY). When defined by a threshold of ≤5 cells/mm³, depletion occurred more rapidly and was more sustained with obinutuzumab. Furthermore, at week 52, depletion persisted in 48% and 94% of patients in LUNAR and NOBILITY, respectively.

Rituximab vs. Obinutuzumab: The same, but different

Type 2 antibodies (Obinutuzumab) differ from type 1 antibodies (rituximab, ofatumumab) by preferentially inducing B-cell depletion through direct cytotoxicity, more so than complement-dependent cytotoxicity (CDC). This is because de-fucosylation of the Fc region of obinutuzumab enhances binding affinity to the Fc?RIII receptor, resulting in enhanced antibody-dependent cell-mediated cytotoxicity (ADCC) and antibody-dependent cell-mediated phagocytosis (ADCP). In contrast, type 1 antibodies (rituximab, ofatumumab) predominantly induce cell death through CDC, because of the ability to redistribute CD20 into lipid rafts.¹²

In both clinical and pre-clinical studies, obinutuzumab has demonstrated more complete and sustained depletion of B-cells than rituximab.^12,13 Even though rituximab and obinutuzumab have not been directly compared head-to-head in a clinical trial, the discordant results between NOBILITY and LUNAR suggest a potential clinical benefit to using obinutuzumab.

More than just peripheral depletion: A closer look at MAINRITSAN2

MAINRITSAN2, a trial of rituximab maintenance therapy in anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis, randomized participants to a “fixed dose” rituximab regimen (500 mg on days 0 and 14, followed by months 6, 12, and 18) or a “tailored dose” rituximab regimen (500 mg on day 0, with re-infusion when CD19⁺ B-lymphocytes exceeded 0 cells/mm³ or ANCA titres increased significantly). The trial found that at 2 years, there was no difference in relapses between the two groups, but the median number of rituximab infusions was reduced in the “tailored dose” group (3 vs. 5).¹⁴

Interestingly however, about half the patients who relapsed had persistent complete peripheral CD19 B-cell depletion, raising the possibility that peripheral B-cell depletion may not always correlate with tissue B-cell depletion.¹⁵

Peripheral B-cell depletion vs. Tissue B-cell depletion

In one study of patients with ANCA vasculitis, ultrasound-guided inguinal lymph node biopsies were performed in both active disease and remission. It was found that even after inducing remission with rituximab, there were a significantly higher number of CD19⁺ B-cells, especially memory B-cells, present within the lymph node compartment compared to the peripheral blood.¹⁶

Tertiary lymphoid tissue in peripheral organs, including the kidneys,¹⁷ may also contribute to the persistence of tissue-resident B-cells, despite apparently complete peripheral depletion.

Given the available data, rituximab depletes B-cells in both peripheral blood and tissues. However, the degree to which both compartments are depleted is likely variable, and the correlation between peripheral B-cell depletion and tissue B-cell depletion is not absolute.

B-cell subsets are differentially affected by anti-CD20 monoclonal antibodies:

Depletion with anti-CD20 monoclonal antibodies affects different subsets of B-cells to different degrees. For example, it is known that rituximab, despite depleting peripheral blood B-cells, has a lesser effect on certain circulating B-cell subsets important in driving disease pathogenesis (such as plasmablasts, double-negative (DN) B-cells, and circulating memory B-cells) and tissue-resident B-cell subsets (such as plasma cells and memory B cells). Notably, obinutuzumab depletes these subsets, potentially contributing to improved therapeutic efficacy. Furthermore, preliminary data suggest that bone marrow plasma cells remain unaffected by obinutuzumab, and thus serum IgG levels do not fall during therapy.¹⁸

CAR-T cell therapy: The ultimate B-cell depletion?

Chimeric antigen receptor (CAR) T-cells are an emerging therapy for resistant lupus nephritis. Vigorous T-cell activation targeting CD19 results in deep, sustained depletion of various B-cell subsets including plasmablasts. A recent case series describes the utility of CD19 CAR-T-cell therapy in eight patients with severe, progressive, and resistant systemic lupus erythematosus. All patients achieved complete remission after a follow-up period ranging from 6 to 29 months, permitting full withdrawal of immunosuppressive therapy. This raises the exciting possibility that CD19 CAR-T cells could reset the B-lymphocyte component of the immune system, suppress autoantibody production, and offer long-term disease remission.¹⁹

Successful clinical use of low-dose rituximab in a clinical setting

Despite the above lines of evidence linking more robust B-cell depletion with clinical outcomes, George et al. have demonstrated that even low-dose rituximab remains efficacious.²⁰ Rituximab (100 mg), in a single dose, was administered to 42 patients with various indications: steroid-dependent nephrotic syndrome (SDNS), frequently relapsing nephrotic syndrome (FRNS), idiopathic MN, and high-immunologic-risk kidney transplantation. Peripheral blood CD19⁺ B-cell counts fell below 1% in 95.2%, 42.9%, and 16.7% of patients by day 30, 90, and 180, respectively, indicating that depletion followed by repopulation had occurred in most cases. However, despite repopulation, clinical remissions were achieved (i.e., 96.7% remission rate for patients with SDNS/FRNS at day 30, and 80% remission rate for patients with MN at 6 months). Although concomitant corticosteroid therapy likely contributed to the remission rates for patients with SDNS/FRNS, rituximab resulted in sustained remissions and lowered corticosteroid requirements.

The flip side: Is less not always better?

A retrospective analysis²¹ was performed to compare two prospective cohorts of PLA2R-associated MN. The NICE cohort (n=28) received two rituximab (1 g) doses, 2 weeks apart, and the GEMRITUX cohort (n=27) received two rituximab (375 mg/m²) doses, 1 week apart. The NICE cohort was found to have higher rates of remission at 6 months, shorter time to achieving remission, better CD19 depletion at 3 months, and lower anti-PLA2R titers at 6 months. This suggests that the higher-dose rituximab protocol may be more effective in achieving remission, which might at least partially be mediated through deeper and longer CD19 depletion.

Another prospective study²² of low-dose rituximab in primary MN compared a single dose (n=18) vs. two doses (n=16) of rituximab (375 mg/m²). Overall, ≤50% of patients had achieved at least a partial response at 12 months. Interestingly, those who did respond had lower baseline anti-PLA2R titers, suggesting that a protocol of low-dose rituximab should not be considered in patients with high anti-PLA2R titers. Traditionally, cyclical cyclophosphamide has been the preferred choice for high-risk MN.³ This recommendation was largely based on data suggesting that rituximab (at doses of 1.5 to 2 g cumulatively) is inferior to cyclical cyclophosphamide in patients with very high anti-PLA2R titres,²³ further reinforcing the notion that lower-dose rituximab should not be considered in these patients.

Dahan et al. have suggested that re-dosing of rituximab at 6 months may be required to achieve remission in cases of insufficient immunological response to rituximab.²⁴ Indeed, this was also the protocol used in the MENTOR trial.²⁵ While re-dosing in both these studies targeted anti-PLA2R titers and not CD19 counts, they add to the evidence that at least some cases of rituximab “resistance” may simply represent inadequate drug dosage (see Supplementary Table S1 for a summary of all the above studies).

While the mechanisms of rituximab resistance are numerous and extensively reviewed elsewhere,²⁶ such non-response due to an insufficient depth and duration of B-cell depletion could be remedied either by a repeat rituximab dose at 6 months,²⁴ or by switching to obinutuzumab.²⁷ Indeed, it has been suggested that in the contemporary era, anti-CD20 therapy should be first-line therapy even in high-risk MN.²⁸

Finally, there is increasing recognition that the development of human anti-chimeric antibodies (HACAs) could contribute to a poor response to rituximab therapy.²⁶ While much remains unknown about the risk factors that contribute to the generation of HACAs, it is possible that repeated exposure to rituximab could play a role. This possibility, though theoretical, again argues against the routine use of repeated low-dose CD19-targeted rituximab administration.

Lowering the rituximab dose has undeniable clinical, logistical, and financial benefits for patients and healthcare systems. Given the inter-individual variability in clinical responses to rituximab, peripheral B-cell depletion, despite its imperfections, continues to be used to assess pharmacodynamic efficacy.

The clinician must remain cognizant of several caveats while using B-cell depletion to guide therapy. Firstly, clinical remissions are strongly associated with complete peripheral B-cell depletion, defined as a CD19 count of 0 cells/mm³. Secondly, peripheral B-cell depletion may be achieved as early as 2 weeks after rituximab in some patients, but delayed responses can be seen for up to a year in others, making it difficult to predict the need for repeat dosing. Thirdly, in addition to the completeness of peripheral blood B-cell depletion, the duration and rapidity of depletion also correlate with clinical outcomes. Finally, even despite achieving complete peripheral B-cell depletion, the persistence of tissue-resident B-cells and certain resistant B-cell subsets may continue to maintain disease activity.

In summary, a “fire-and-forget” strategy of administering “full/standard-dose” rituximab continues to be challenged by the more precision-medicine approach of a “treat-to-target” strategy using B-cell depletion to guide re-dosing. However, there is still much nuance in defining the target: depth and duration of B-cell depletion.